Sampling strategy

There are two types of samples that can be collected in a sampling event grab samples and composite samples. 

Grab samples are discrete aliquots representing a specific location at a specific time 

Composite samples are typically collected for soil stockpile or surface area characterization. Twenty-four-hour composite samples of water may be collected with automated composite samplers from streams or process piping composite samples may be made of several grab samples collected from different depths in a soil boring. 

Probabilistic sampling, which lies in the core of the DQO process, is based on a random sample location selection strategy. It produces data that can be statistically evaluated in terms of the population mean, variance, standard deviation, standard error, confidence interval, and other statistical parameters. The EPA provides detailed guidance on the DQO process application for the 

Depending on the variability of the sampled medium and the purpose of sampling, several types of probabilistic sampling designs may be used. 

Generation of test atmospheres of organic vapours by the syringe injection technique Generation of test atmospheres of organic vapours by the permeation tube method On-site validation of sampling methods 

General methods for sampling and gravimetric analysis of respirable and inhalable dust 

Sorbent tube standards preparation by the syringe injection technique 

Protocol for assessing the performance of a pumped sampler for gases and vapours 

General methods for sampling airborne gases and vapours 

On land contaminated by industrial activities homogeneously distributed contamination will be the exception rather than the rule and contamination will in general be heterogeneous both horizontally and vertically over the site in question. The strategies appropriate to soil sampling in such circumstances have recently been discussed (Fortunati et al., 1994). 

Two approaches are available. In the first a systematic survey is made of samples from a grid of say 50 m2 at a fixed depth of 50 cm (Fortunati et al., 1994). These samples are inspected and the appropriate parameters measured. The results of this preliminary study act as a guide to further sampling and analysis. With this type of systematic sampling plan the number of samples can easily become unmanageable and their study uneconomic. Furthermore, sampling at a fixed depth will not be useful where the site has been subject to landfill or, as is often the case on industrial sites, has been built up by deposits of fill material itself contaminated. 

10 X sensitivity of analytical method (mg) suitable hygiene standard (mg/m ) 

For routine monitoring, frequency will be influenced by the level of exposure. The further the levels depart from the standard, the less the need for routine monitoring. A suggested guide is  

Routine monitoring becomes superfluous if exposures are very low (i.e. below 0.1 x hygiene standard) or very high (e.g. 10 x hygiene standard), although where standards 

Preliminary All 0.1 XOEL 0.1-1.5XOEL Some 1.5 X OEL None if exposures are as low as reasonably practical. Investigate and carry out a more detailed survey. Investigate, take remedial action and repeat survey. 

Generation of test atmospheres of organic vapours by the permeation tube method 4 

---

Both sampling strategies require a total of ten determinations. Using Appendix IB, we find that the value of t is 2.26. Substituting into equation 7.8, we find that the relative error for the first sampling strategy is... 

As expected, since the relative method variance is better than the relative sampling variance, a sampling strategy that favors the collection of more samples and few replicate analyses gives the better relative error. 

Measurement Method Selection. A measurement method should meet sampling strategy requirements to the degree that the data can be used for decision making. This does not mean that it must be the optimum method with respect to all requirements. The range of methods available is limited and it is often necessary to select a method deficient in one or more attributes but which can yield data from which conclusions can be drawn with the desired degree of confidence. Some of the attributes to be considered in selecting a method foUow. 

N. A. Leidel, K. A, Busch, and J. R. Lynch, Occupational Exposure Sampling Strategy Manual, pubhcation no. 77-173, U.S. Department of Health and Human Services (NIOSH), Washington, D.C., 1977. 

A straightforward Gibbs sampling strategy when the number of components is known (or fixed) is as follows [48]. 

Sampling strategies may include measurement of the hazardous substance ... 

For a few substances exposure may be assessed using biological monitoring (see page 114). Depending upon the substance the sampling strategy varies from post shift, random, or pre-shift the day after exposure. 

At least three factors must be considered when using Eq. (10.132). The first is that the room air is seldom well mixed the second is that the emission rate is usually unknown and the third is that measurement of is often very difficult, especially in spaces where a mixture of mechanical and natural ventilation is used. If the room air is poorly mixed, a sampling strategy is needed in order to determine the room air concentration (C,). There exist different strategies for this, and two are as follows ... 

One of the most important elements in the remediation of existing waste sites is early detection and actiom As an example, the cost of cleanup at Stringfellow, California, increased from an estimated 3.4 million to 65 million because of pollutant dispersal during a decade of inaction after the first identification of the problem. The opportunities for innovative sampling strategies responsive to this need are discussed in the following section. 

These direct sampling strategies provide estimators that are... 

The administrators or users of the study results must supply the objectives and required precision. Statisticians can develop the models for alternative sampling strategies. The estimates of variance components and costs can come from a number of places ... 

The sampling strategy outlined is designed to provide 95% confidence that the average emission rate estimate (for a given zone) will be within 20% of the true mean. 

---